CN109712398B - Expressway travel time estimation model parameter optimization method - Google Patents
Expressway travel time estimation model parameter optimization method Download PDFInfo
- Publication number
- CN109712398B CN109712398B CN201910060915.4A CN201910060915A CN109712398B CN 109712398 B CN109712398 B CN 109712398B CN 201910060915 A CN201910060915 A CN 201910060915A CN 109712398 B CN109712398 B CN 109712398B
- Authority
- CN
- China
- Prior art keywords
- travel time
- time estimation
- model
- estimation model
- road section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a parameter optimization method of a highway travel time estimation model, which comprises the steps of constructing a route travel time estimation model comprising a plurality of road sections based on a road section travel time estimation model; optimizing model parameters l, m and alpha by adopting a genetic algorithm; defining a genetic algorithm fitness function
Wherein Wj is an error index MAjJ is the time interval of the travel time estimation, and the value range is [2, K ]]The optimization goal of the genetic algorithm is that the fitness function is minimum; calculating the weight Wj by using an M-H algorithm; outputting the optimal parameter combination [ l, m, alpha ] under the free flow state and the congestion state by a genetic algorithm](ii) a Further establishing an optimized route travel time estimation model; according to the method, the sensitivity of the model to different traffic running states such as congestion and smoothness is improved by automatically optimizing the parameters of the travel time estimation model, the travel time estimation precision under the congestion condition is obviously improved, and the reliability and stability of the model performance are guaranteed.
Description
Technical Field
The invention relates to a method for optimizing parameters of an expressway travel time estimation model.
Background
The travel time is one of important indexes for evaluating traffic operation congestion conditions, and the travel time can provide data reference for implementation of a traffic guidance scheme and release of travel information during traffic guidance, and is the most intuitive basis for travelers to make travel decisions.
The current travel time parameters are acquired indirectly through a certain data processing means based on traffic flow parameters acquired by fixed detection equipment. The expressway travel time estimation model can be divided into a track-based estimation model, a vehicle identification model and a traffic flow theoretical model.
However, many current travel time estimation models have the problem of model performance stability, and particularly under congested road conditions, the accuracy of travel time estimation is low. The above-mentioned problem is a problem that should be considered and solved in the highway travel time estimation process.
Disclosure of Invention
The invention aims to provide a method for optimizing parameters of a highway travel time estimation model, which ensures that the model has good estimation precision in different running states, and solves the problems that the current travel time estimation model in the prior art has model performance stability, and particularly the travel time estimation precision is low under congested road conditions.
The technical solution of the invention is as follows:
a method for optimizing the parameters of the travel time estimation model of expressway includes such steps as,
s1, respectively setting virtual following vehicles VF and front vehicles VL at node positions Su and Sd of upstream and downstream detectors of a road section, constructing a GM following model, describing the operation parameter relationship between the front vehicles and the following vehicles, and constructing a road section travel time estimation model GMTTE on the basis:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity
Wherein v isSd、vSuAs input values for the model, respectively for the section IthA speed detection value output by the upstream and downstream detectors; l is the locomotive spacing index, m is the speed index, and alpha is the sensitivity; further constructing a route travel time estimation model comprising a plurality of road segments based on the road segment travel time estimation model;
the values of the S2, l, m and alpha parameters directly influence the travel time estimation precision, and a genetic algorithm fitness function is predefined
Wherein Wj is an error index MAjJ is the time interval of the travel time estimation, and the value range is [2, K ]]The optimization target of the genetic algorithm is the minimum fitness function;
s3, initializing the weight of the genetic algorithm fitness function, namely p (j) ═ Wj, and determining p (j) prior distribution according to the traffic flow state; calculating error index MAj(l, m, alpha), updating the weight by adopting a Metropolis-Hastings algorithm, iteratively calculating the optimal solution of the time interval j of travel time estimation through a genetic algorithm, and determining the corresponding weight Wj;
s4, based on the optimal solution corresponding weight determined by S3, outputting an optimal parameter combination [ l, m, alpha ] in a free flow state and a congestion state by a genetic algorithm, and optimizing the initial route travel time estimation model constructed by S1.
Further, in step S1, a route travel time estimation model including a plurality of road segments is constructed, specifically:
s11, using the distribution point of the highway traffic flow detector as the road segment dividing node in the route, numbering the road segments (I + k)thI is a route number, k is a road section serial number, th is a road section number mark, and k belongs to [0, n ]]I.e. the route is divided into n +1 road segments;
s12, for any road section (I + k)thVirtual following vehicles VF and virtual leading vehicles VL are respectively arranged at the upstream detector node position Su and the downstream detector node position Sd; the real-time running speed of VL is calculated according to a speed detection value output by a downstream detector node position Sd detector, and the formula is as follows:
where T ∈ [ T ]i,Ti+1]，TiA data timestamp for the downstream detector node location Sd detector,
respectively at time stamp T for the detectori、Ti+1A running speed detection value of (1);
s13, constructing a road section travel time estimation model GMTTE according to the relation between the GM following model and the operation parameters of the leading vehicle and the following vehicle:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity
Wherein v isSd、vSuAs input values for the model, respectively for the section IthThe speed detection values output by the upstream detector and the downstream detector; l is the locomotive spacing index, m is the speed index, and alpha is the sensitivity; on this basis, the route travel time is estimated.
Further, in step S13, the route travel time is estimated, specifically,
s131, following vehicle VF on road section IthReal-time running speed at any time t of up running
The calculation formula is as follows:
in the formula
The running speed and the acceleration of VF at the time of T-delta T and the real-time acceleration
The specific calculation formula is as follows:
in the formula
The running speed of the lead vehicle at the moment T-delta T,
respectively the running distance of the leading vehicle and the following vehicle at the time of T-delta T and the real-time running distance of the VL of the leading vehicle
The calculation formula is as follows:
real-time driving distance of following vehicle VF
The calculation formula is as follows:
in the formula
The running speed of the lead vehicle at the moment T-delta T,
for the running distance of the following vehicle at the time T-delta T,
the acceleration of VF at the time T-delta T;
s132, taking the road section IthEnd point real time velocity vTTIFor following car VF on the next road section (I +1)thInitializing the speed of the starting point;
s133, estimating model by using road section travel time
Estimation section (I +1)thThe travel time of (a);
s134, circulating the steps S132 and S133 until the road section (I + n) is finishedthA travel time estimate of (d);
s135, route travel time
Wherein TTIFor following vehicle VF on road section IthUpper travel time.
Further, in step S3, specifically,
s31, initializing weights, i.e., p (j) ═ Wj; the prior distribution p (j) is selected according to the traffic flow running state, and the traffic flow running state comprises a free flow state and a congestion state;
s32, calculating an error index MAj(l, m, α), adding MAjThe minimum value of the corresponding j is taken as the optimal time interval; the genetic algorithm outputs an optimal solution every iteration
Wherein M is the number of iterations;
s33, passing the optimal solution
And updating the prior distribution in the two operation states in the step S31 to generate the posterior distribution of the optimal solution.
Further, in step S31, the prior distribution p (j) is selected according to the traffic flow running state, specifically,
using uniform probability distribution in free-flow regime, i.e.
σ1、σ2The upper and lower limits of j are defined according to the definition of the value range in S2, σ1＝2，σ2＝K；
Using lognormal distribution in congested conditions, i.e.
μ and σ are the mean and standard deviation of j, respectively.
Further, in step S33, specifically,
s331, representing the known optimal solution j by Gaussian distributionoldAnd iterating the newly generated optimal solution jnewThe probability relationship between p (j)new|jold)＝N(jold,σr) Wherein j isoldThe initial value is set to a non-negative number, σrIs the step length;
s332, calculating likelihood ratio r, and judging jnewWhether it is acceptable; in the free flow state, the calculation formula of r is as follows:
in the congestion state, the calculation formula of r is as follows:
if r > 1, then jnewAcceptable, the current optimal solution is calculated by joldIs updated to jnewOtherwise, it remains as jold；
S333, performing N-fold loop on the steps S331 and S332 to obtain the optimal solution J of the time interval J and the corresponding weight WJ。
The invention has the beneficial effects that:
according to the method for optimizing the parameters of the expressway travel time estimation model, the sensitivity of the model to different traffic running states such as congestion and unblocked traffic is improved by automatically optimizing the parameters of the travel time estimation model, the travel time estimation precision under the congestion condition is obviously improved, and the reliability and stability of the model performance are guaranteed.
The method for optimizing the parameters of the expressway travel time estimation model estimates the travel time of the road section based on the GM following model, and then establishes a route travel time estimation model by applying a continuous speed model on the basis of the estimation of the travel time of the expressway, so that the estimation of the travel time of the route is realized.
The method for optimizing the parameters of the expressway travel time estimation model fully utilizes distributed computing performance, optimizes the parameters of the route travel time estimation model by adopting a parallel genetic algorithm and a Metropolis-Hastings algorithm, fully considers the influence of different congestion levels on the travel time estimation model, and further improves the accuracy of the model in forming estimation on the congested and unblocked routes in different running states.
Drawings
FIG. 1 is a flow chart of a method for optimizing parameters of a highway travel time estimation model according to an embodiment of the invention.
Fig. 2 is a schematic diagram of setting the virtual leading vehicle VL and the following vehicle VF in the embodiment.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
Examples
According to the method for optimizing the parameters of the expressway travel time estimation model, a road section GMTTE (General Motor's travel time estimation) travel time estimation model is combined with a continuous speed model, the estimation of the route travel time is realized, the distributed computing performance is fully utilized, the parameters of the route travel time estimation model are optimized by adopting a parallel genetic algorithm, the influence of different congestion levels on the travel time estimation model is fully considered, and the accuracy of the model in estimation of congested and unblocked routes in different running states is further improved.
A method for optimizing parameters of an expressway travel time estimation model, as shown in figure 1, comprises the following steps,
s1, respectively setting virtual following vehicles VF and front vehicles VL at node positions Su and Sd of upstream and downstream detectors of a road section, constructing a GM following model, describing the operation parameter relationship between the front vehicles and the following vehicles, and constructing a road section travel time estimation model GMTTE on the basis:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity
Wherein v isSd、vSuAs input values for the model, respectively for the section IthA speed detection value output by the upstream and downstream detectors; l is the locomotive spacing index, m is the speed index, and alpha is the sensitivity; further constructing a route travel time estimation model comprising a plurality of road segments based on the road segment travel time estimation model; the method comprises the following specific steps:
s11, using the distribution point of the highway traffic flow detector as the road segment dividing node, and numbering the road segments (I + k)thI is a route number, k is a road section serial number, th is a road section number mark, and k belongs to [0, n ]]I.e. the route is divided into n +1 road segments.
S12, for any road section (I + k)thVirtual following vehicles VF and virtual leading vehicles VL are respectively arranged at the upstream detector node position Su and the downstream detector node position Sd, as shown in fig. 2; the real-time running speed of VL is calculated according to a speed detection value output by a node position Sd detector of a downstream detector, and the formula is as follows:
where T ∈ [ T ]i,Ti+1]，TiIs the data time stamp of the Sd detector,
respectively at time stamp T for the detectori、Ti+1The detected average traveling speed.
S13, according to the definition of the GM car following model on the operation parameter relation between the leading car and the car following, constructing a road section travel time estimation model GMTTE:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity vTTIWherein v isSd、vSuAs input values for the model, respectively for the section IthThe speed detection values output by the upstream detector and the downstream detector; l is the locomotive spacing index, m is the speed index, and alpha is the sensitivity; the process is repeated until the road segment (I + n) is completedthEstimating the travel time of the route, and further estimating the travel time of the route; in particular, the amount of the solvent to be used,
s131, following vehicle VF on road section IthReal-time running speed at any time t of up runningDegree of rotation
The calculation formula is as follows:
in the formula
The running speed and the acceleration of VF at the time of T-delta T and the real-time acceleration
The specific calculation formula is as follows:
wherein a (—) is the sensitivity,
the running speed of the lead vehicle at the moment T-delta T,
respectively the running distance of the leading vehicle and the following vehicle at the time of T-delta T and the real-time running distance of the VL of the leading vehicle
The calculation formula is as follows:
real-time driving distance of following vehicle VF
The calculation formula is as follows:
in the formula
The running speed of the lead vehicle at the moment T-delta T,
for the running distance of the following vehicle at the time T-delta T,
the acceleration of VF at time T- Δ T.
S132, taking the road section IthEnd point real time velocity vTTIFor following car VF on the next road section (I +1)thThe speed of the starting point is initialized.
S133, estimating model by using road section travel time
Estimation section (I +1)thThe travel time of (c).
S134, circulating the steps S132 and S133 until the road section (I + n) is finishedthIs estimated.
S135, route travel time
Wherein TTIFor following vehicle VF on road section IthUpper travel time.
The values of the S2, l, m and alpha parameters directly influence the travel time estimation precision, and a genetic algorithm fitness function is predefined
Wherein Wj is an error index MAjJ is the time interval of the travel time estimation, and the value range is [2, K ]]The time unit is minute (min), in the embodiment, the maximum estimation interval K of the travel time is 15, that is, the upper and lower limits of the estimation interval of the travel time are 2min and 15min respectively; wherein, the error index MA is selected from the average absolute error MAE and the average absolute percentage error MAPE; the genetic algorithm optimization goal is that the fitness function is minimal.
S3, initializing the weight of the genetic algorithm fitness function, namely p (j) ═ Wj, and determining p (j) prior distribution according to the traffic flow state; calculating error index MAj(l, m, α) using Metropolis-HastingsUpdating the weight by the method, iteratively calculating the optimal solution of the time interval j of the travel time estimation through a genetic algorithm, and determining the corresponding weight Wj; the specific calculation method comprises the following steps:
s31, initializing weights, i.e., p (j) ═ Wj; wherein the prior distribution p (j) is selected according to the traffic flow running state and adopts uniform probability distribution in the free flow state, i.e.
σ1、σ2Respectively, the upper and lower value limits of j, according to the definition of the value range of the j in the step S2, sigma1＝2，σ215; using lognormal distribution in congested conditions, i.e.
Wherein μ and σ are the mean value and standard deviation of j, μ is 3, and σ is 1, respectively.
S32, calculating an error index MAj(l, m, α), adding MAjThe minimum value of the corresponding j is taken as the optimal time interval; the genetic algorithm outputs an optimal solution every iteration
Where M is the number of iterations.
S33, passing the optimal solution
Updating prior distributions in the two running states of S31 to generate posterior distribution of an optimal solution; the method comprises the following specific steps:
s331, representing the known optimal solution j by Gaussian distributionoldAnd iterating the newly generated optimal solution jnewThe probability relationship between p (j)new|jold)＝N(jold,σr) Wherein j isoldThe initial value is set to a non-negative number, in the embodiment 10, step size σrThe value is 5;
s332, calculating likelihood ratio r, and judging jnewWhether it is acceptable; in the free flow state, the calculation formula of r is as follows:
in the congestion state, the calculation formula of r is as follows:
if r > 1, then jnewAcceptable, the current optimal solution is calculated by joldIs updated to jnewOtherwise, it remains as jold；
S333, performing N-fold circulation on S331 and S332 to obtain an optimal solution J of a time interval J and a corresponding weight Wi(ii) a Wherein, in the embodiment, the value of N is 2000.
S4, outputting an optimal parameter combination [ l, m, alpha ] in a free flow state and a congestion state by a genetic algorithm based on the optimal solution corresponding weight determined in the step S3; the initial route travel time estimation model constructed in step S1 is optimized.
According to the method for optimizing the parameters of the expressway travel time estimation model, the parameters of the travel time estimation model are automatically optimized, so that the sensitivity of the model to different traffic running states such as congestion and unblocked traffic is improved, the travel time estimation precision under the congestion condition is obviously improved, and the reliability and stability of the model performance are guaranteed.
The method for optimizing the parameters of the expressway travel time estimation model estimates the travel time of the road section based on the GM following model, and on the basis, a continuous speed model is applied to construct a route travel time estimation model to realize the estimation of the route travel time. According to the method for optimizing the parameters of the expressway travel time estimation model, the parameters of the route travel time estimation model are automatically optimized by adopting a genetic algorithm and a Metropolis-Hastings algorithm, and the reliability of travel time estimation performance in different running states of traffic flow is improved.
One specific example of an embodiment is as follows:
the method comprises the steps of adopting an optimization model optimized and constructed by the method of the embodiment to estimate the travel time of two expressway routes, wherein the total length of the route is 11.59km, the number of lanes is 3, the selected analysis time period is 7: 00-8: 00, the traffic flow is mostly in a smooth state in the time period, the total length of the route is 5.95km, the number of lanes is 4, the analysis time period is 7: 50-8: 50, and congestion occurs frequently in the time period. 183 pieces and 285 pieces of real data of the travel time of the two lines are obtained respectively to calculate estimation errors.
In the embodiment, when the genetic algorithm definition is performed in step S2, the mean absolute error MAE and the mean absolute percentage error MAPE are respectively used as error indicators to establish a fitness function, so as to perform parameter optimization, and the final model parameter optimization result is as shown in the following table.
Taking the MAPE as an error index, the road section travel time estimation model GMTTE of the route 1 is as follows:
section travel time estimation model GMTTE of route 2:
according to
Estimating the route travel time; with MAE as an error index, a road section travel time estimation model GMTTE of the route 1 is as follows:
section travel time estimation model GMTTE of route 2:
according to
A route travel time estimate is made.
Claims (6)
1. A method for optimizing parameters of an expressway travel time estimation model is characterized by comprising the following steps: comprises the following steps of (a) carrying out,
s1, upper and lower road sectionsVirtual following vehicles VF and virtual leading vehicles VL are respectively arranged at the node positions Su and Sd of the trip detectors, a GM following model is constructed, the operation parameter relation between the leading vehicles and the following vehicles is described, and a section travel time estimation model GMTTE is constructed on the basis:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity
Wherein v isSd、vSuAs input values for the model, respectively for the section IthA speed detection value output by the upstream and downstream detectors; l is the locomotive spacing index, m is the speed index, and alpha is the sensitivity; further constructing a route travel time estimation model comprising a plurality of road segments based on the road segment travel time estimation model;
the values of the S2, l, m and alpha parameters directly influence the travel time estimation precision, and a genetic algorithm fitness function is predefined
Wherein Wj is an error index MAjJ is the time interval of the travel time estimation, and the value range is [2, K ]]The optimization target of the genetic algorithm is the minimum fitness function;
s3, initializing the weight of the genetic algorithm fitness function, namely p (j) ═ Wj, and determining p (j) prior distribution according to the traffic flow state; calculating error index MAj(l, m, alpha), updating the weight by adopting a Metropolis-Hastings algorithm, iteratively calculating the optimal solution of the time interval j of travel time estimation through a genetic algorithm, and determining the corresponding weight Wj;
s4, based on the optimal solution corresponding weight determined in the step S3, outputting an optimal parameter combination [ l, m, alpha ] in a free flow state and a congestion state by a genetic algorithm, and optimizing the initial route travel time estimation model constructed in the step S1.
2. The method of optimizing highway travel time estimation model parameters according to claim 1, wherein: in step S1, a route travel time estimation model including a plurality of road segments is constructed, specifically:
s11, using the distribution point of the highway traffic flow detector as the road segment dividing node in the route, numbering the road segments (I + k)thI is a route number, k is a road section serial number, th is a road section number mark, and k belongs to [0, n ]]I.e. the route is divided into n +1 road segments;
s12, for any road section (I + k)thVirtual following vehicles VF and virtual leading vehicles VL are respectively arranged at the upstream detector node position Su and the downstream detector node position Sd; the real-time running speed of VL is calculated according to a speed detection value output by a downstream detector node position Sd detector, and the formula is as follows:
where T ∈ [ T ]i,Ti+1]，TiA data timestamp for the downstream detector node location Sd detector,
respectively at time stamp T for the detectori、Ti+1The detected average traveling speed;
s13, according to the definition of the GM car following model on the operation parameter relation between the leading car and the car following, constructing a road section travel time estimation model GMTTE:
the model outputs a virtual following vehicle VF on a road section IthUpper travel time TTIAnd on the road section IthDownstream endpoint real time velocity
Wherein v isSd、vSuAs input values for the model, respectively for the section IthThe speed detection values output by the upstream detector and the downstream detector; l is a vehicleHead spacing index, m velocity index, α sensitivity; the process is repeated until the road segment (I + n) is completedthAnd estimating the route travel time.
3. The method for optimizing highway travel time estimation model parameters according to claim 2, wherein: in step S13, the route travel time is estimated, specifically,
s131, following vehicle VF on road section IthReal-time running speed at any time t of up running
The calculation formula is as follows:
in the formula
The running speed and the acceleration of VF at the time of T-delta T and the real-time acceleration
The specific calculation formula is as follows:
wherein a (—) is the sensitivity,
the running speed of the lead vehicle at the moment T-delta T,
respectively the running distance of the leading vehicle and the following vehicle at the time of T-delta T and the real-time running distance of the VL of the leading vehicle
The calculation formula is as follows:
real-time driving distance of following vehicle VF
The calculation formula is as follows:
in the formula
The running speed of the lead vehicle at the moment T-delta T,
for the running distance of the following vehicle at the time T-delta T,
the acceleration of VF at the time T-delta T;
s132, taking the road section IthEnd point real time velocity vTTIFor following car VF on the next road section (I +1)thInitializing the speed of the starting point;
s133, estimating model by using road section travel time
Estimation section (I +1)thThe travel time of (a);
s134, circulating the steps S132 and S133 until the road section (I + n) is finishedthA travel time estimate of (d);
s135, route travel time
Wherein TTIFor following vehicle VF on road section IthUpper travel time.
4. The highway travel time estimation model parameter optimization method of any one of claims 1-3, wherein: in step S3, specifically, the step,
s31, initializing weights, i.e., p (j) ═ Wj; the prior distribution p (j) is selected according to the traffic flow running state, and the traffic flow running state comprises a free flow state and a congestion state;
s32, calculating an error index MAj(l, m, α), adding MAjThe minimum value of the corresponding j is taken as the optimal time interval; the genetic algorithm outputs an optimal solution every iteration
Wherein M is the number of iterations;
s33, passing the optimal solution
And updating the prior distribution in the two operation states in the step S31 to generate the posterior distribution of the optimal solution.
5. The method of optimizing highway travel time estimation model parameters according to claim 4, wherein: in step S31, the prior distribution p (j) is selected according to the traffic flow running state, specifically,
using uniform probability distribution in free-flow regime, i.e.
σ1、σ2The upper and lower limits of j are defined according to the definition of the value range in S2, σ1＝2，σ2＝K；
Using lognormal distribution in congested conditions, i.e.
μ and σ are the mean and standard deviation of j, respectively.
6. The method of optimizing highway travel time estimation model parameters according to claim 4, wherein: in step S33, specifically, the step,
s331, use highThe gaussian distribution represents the known optimal solution joldAnd iterating the newly generated optimal solution jnewThe probability relationship between p (j)new|jold)＝N(jold,σr) Wherein j isoldThe initial value is set to a non-negative number, σrIs the step length;
s332, calculating likelihood ratio r, and judging jnewWhether it is acceptable; in the free flow state, the calculation formula of r is as follows:
in the congestion state, the calculation formula of r is as follows:
if r > 1, then jnewAcceptable, the current optimal solution is calculated by joldIs updated to jnewOtherwise, it remains as jold；
S333, performing N-fold loop on the steps S331 and S332 to obtain the optimal solution J of the time interval J and the corresponding weight WJ。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910060915.4A CN109712398B (en) | 2019-01-22 | 2019-01-22 | Expressway travel time estimation model parameter optimization method |
| PCT/CN2019/115052 WO2020151293A1 (en) | 2019-01-22 | 2019-11-01 | Parameter optimization method for expressway travel time estimation model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910060915.4A CN109712398B (en) | 2019-01-22 | 2019-01-22 | Expressway travel time estimation model parameter optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109712398A CN109712398A (en) | 2019-05-03 |
| CN109712398B true CN109712398B (en) | 2021-05-28 |
Family
ID=66262603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910060915.4A Active CN109712398B (en) | 2019-01-22 | 2019-01-22 | Expressway travel time estimation model parameter optimization method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109712398B (en) |
| WO (1) | WO2020151293A1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712398B (en) * | 2019-01-22 | 2021-05-28 | 江苏智通交通科技有限公司 | Expressway travel time estimation model parameter optimization method |
| CN110322054B (en) * | 2019-06-14 | 2023-04-28 | 中交第一公路勘察设计研究院有限公司 | Optimized layout method of road section traffic monitor |
| CN110634292B (en) * | 2019-09-19 | 2020-10-16 | 北京航空航天大学 | Travel time reliability estimation method based on road resistance performance function |
| CN111125862B (en) * | 2019-09-27 | 2023-12-26 | 长安大学 | Following model emission measuring and calculating method based on genetic algorithm and specific power |
| CN113379233B (en) * | 2021-06-08 | 2023-02-28 | 重庆大学 | Travel time reliability estimation method and device based on high-order moment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0877485A (en) * | 1994-09-08 | 1996-03-22 | Toshiba Corp | Predicting device for required travel time |
| CN101794507A (en) * | 2009-07-13 | 2010-08-04 | 北京工业大学 | Method for evaluating macroscopic road network traffic state based on floating car data |
| CN102991498A (en) * | 2011-12-19 | 2013-03-27 | 王晓原 | Driver following behavior model based on multi-source information fusion |
| CN106384507A (en) * | 2016-09-20 | 2017-02-08 | 宁波大学 | Travel time real-time estimation method based on sparse detector |
| CN106652458A (en) * | 2017-02-20 | 2017-05-10 | 东南大学 | Online urban road route travel time estimation method based on virtual vehicle locus reconstruction |
| CN106971536A (en) * | 2017-04-17 | 2017-07-21 | 武汉云砥信息科技有限公司 | It is a kind of to merge Floating Car and the highway congestion status detection method of trackside video |
| CN107554524A (en) * | 2017-09-12 | 2018-01-09 | 北京航空航天大学 | A kind of following-speed model stability control method based on subjective dangerous criminal |
| CN108417029A (en) * | 2018-02-11 | 2018-08-17 | 东南大学 | City road network travel time estimation method based on adaptive multitask deep learning |
| CN109155104A (en) * | 2016-04-18 | 2019-01-04 | 北京嘀嘀无限科技发展有限公司 | Recommend the system and method for Estimated Time of Arrival |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712398B (en) * | 2019-01-22 | 2021-05-28 | 江苏智通交通科技有限公司 | Expressway travel time estimation model parameter optimization method |
-
2019
- 2019-01-22 CN CN201910060915.4A patent/CN109712398B/en active Active
- 2019-11-01 WO PCT/CN2019/115052 patent/WO2020151293A1/en active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0877485A (en) * | 1994-09-08 | 1996-03-22 | Toshiba Corp | Predicting device for required travel time |
| CN101794507A (en) * | 2009-07-13 | 2010-08-04 | 北京工业大学 | Method for evaluating macroscopic road network traffic state based on floating car data |
| CN102991498A (en) * | 2011-12-19 | 2013-03-27 | 王晓原 | Driver following behavior model based on multi-source information fusion |
| CN109155104A (en) * | 2016-04-18 | 2019-01-04 | 北京嘀嘀无限科技发展有限公司 | Recommend the system and method for Estimated Time of Arrival |
| CN106384507A (en) * | 2016-09-20 | 2017-02-08 | 宁波大学 | Travel time real-time estimation method based on sparse detector |
| CN106652458A (en) * | 2017-02-20 | 2017-05-10 | 东南大学 | Online urban road route travel time estimation method based on virtual vehicle locus reconstruction |
| CN106971536A (en) * | 2017-04-17 | 2017-07-21 | 武汉云砥信息科技有限公司 | It is a kind of to merge Floating Car and the highway congestion status detection method of trackside video |
| CN107554524A (en) * | 2017-09-12 | 2018-01-09 | 北京航空航天大学 | A kind of following-speed model stability control method based on subjective dangerous criminal |
| CN108417029A (en) * | 2018-02-11 | 2018-08-17 | 东南大学 | City road network travel time estimation method based on adaptive multitask deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020151293A1 (en) | 2020-07-30 |
| CN109712398A (en) | 2019-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109712398B (en) | Expressway travel time estimation model parameter optimization method | |
| Hourdakis et al. | Practical procedure for calibrating microscopic traffic simulation models | |
| Cortes et al. | General-purpose methodology for estimating link travel time with multiple-point detection of traffic | |
| CN107293115B (en) | Traffic flow prediction method for microscopic simulation | |
| JP5901838B2 (en) | How to predict future travel time on a link | |
| CN107610464A (en) | A kind of trajectory predictions method based on Gaussian Mixture time series models | |
| CN107945507A (en) | Travel Time Estimation Method and device | |
| JP2011022649A (en) | Link travel time calculation device and program | |
| CN107111938B (en) | Server, system and method for determining congestion end position | |
| CN109410587A (en) | A kind of Macro-traffic Flow method for parameter estimation of city expressway | |
| CN104408924A (en) | Detection method for abnormal traffic flow of urban road based on coupled hidden markov model | |
| CN105006149B (en) | Traffic estimates Dynamic iterations method | |
| CN114450557A (en) | Route deviation quantification and vehicle route learning based thereon | |
| CN110909907A (en) | Method and device for predicting fuel consumption of truck and storage medium | |
| CN114299742A (en) | Dynamic recognition and updating recommendation method for speed limit information of expressway | |
| CN104169986A (en) | Method for model construction for a travel-time database | |
| JP2012048444A (en) | Traffic information processing device, traffic information processing system, program, and traffic information processing method | |
| JP4030848B2 (en) | Traffic situation estimation device and OD traffic volume correction device | |
| Jayasinghe et al. | Calibration of SUMO microscopic simulator for Sri Lankan traffic conditions | |
| CN105116373B (en) | Target IP region city-class positioning algorithm based on indirect time delay | |
| JP2015046186A (en) | Traffic information processing device, traffic information processing system, program, and traffic information processing method | |
| CN111160594B (en) | Method and device for estimating arrival time and storage medium | |
| CN117012028A (en) | Signal lamp countdown prediction method based on multi-source data fusion | |
| CN116989801A (en) | Map matching method and device for low-frequency long tracks of complex road network | |
| CN111613052A (en) | Traffic condition determining method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 211106 19 Su Yuan Avenue, Jiangning economic and Technological Development Zone, Nanjing, Jiangsu Applicant after: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. Address before: 210006, Qinhuai District, Jiangsu, Nanjing should be 388 days street, Chenguang 1865 Technology Creative Industry Park E10 building on the third floor Applicant before: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |